An aquaculture-based method for calibrated bivalve isotope paleothermometry Alan D. Wanamaker Jr., Karl J. Kreutz, Harold W. Borns Jr., and Douglas S. Introne Climate Change Institute and Department of Earth Sciences, University of Maine, 5790 Bryand Global Sciences Center, Orono, Maine 04469, USA ([email protected]) Scott Feindel Darling Marine Center, University of Maine, Walpole, Maine 04573, USA Bruce J. Barber School of Marine Sciences, University of Maine, Orono, Maine 04469, USA Now at Galbraith Marine Science Laboratory, Eckerd College, St. Petersburg, Florida 33711, USA [1] To quantify species-specific relationships between bivalve carbonate isotope geochemistry (d 18 O c ) and water conditions (temperature and salinity, related to water isotopic composition [d 18 O w ]), an aquaculture- based methodology was developed and applied to Mytilus edulis (blue mussel). The four-by-three factorial design consisted of four circulating temperature baths (7, 11, 15, and 19°C) and three salinity ranges (23, 28, and 32 parts per thousand (ppt); monitored for d 18 O w weekly). In mid-July of 2003, 4800 juvenile mussels were collected in Salt Bay, Damariscotta, Maine, and were placed in each configuration. The size distribution of harvested mussels, based on 105 specimens, ranged from 10.9 mm to 29.5 mm with a mean size of 19.8 mm. The mussels were grown in controlled conditions for up to 8.5 months, and a paleotemperature relationship based on juvenile M. edulis from Maine was developed from animals harvested at months 4, 5, and 8.5. This relationship [T°C = 16.19 (±0.14) 4.69 (±0.21) {d 18 O c VPBD d 18 O w VSMOW} + 0.17 (±0.13) {d 18 O c VPBD d 18 O w VSMOW} 2 ;r 2 = 0.99; N = 105; P < 0.0001] is nearly identical to the Kim and O’Neil (1997) abiogenic calcite equation over the entire temperature range (7–19°C), and it closely resembles the commonly used paleotemperature equations of Epstein et al. (1953) and Horibe and Oba (1972). Further, the comparison of the M. edulis paleotemperature equation with the Kim and O’Neil (1997) equilibrium-based equation indicates that M. edulis specimens used in this study precipitated their shell in isotopic equilibrium with ambient water within the experimental uncertainties of both studies. The aquaculture-based methodology described here allows similar species-specific isotope paleothermometer calibrations to be performed with other bivalve species and thus provides improved quantitative paleoenvironmental reconstructions. Components: 7964 words, 4 figures, 1 table. Keywords: paleothermometry; aquaculture methods; bivalves; isotope geochemistry; sea surface temperature proxy; paleoceanography. Index Terms: 4215 Oceanography: General: Climate and interannual variability (1616, 1635, 3305, 3309, 4513); 4870 Oceanography: Biological and Chemical: Stable isotopes (0454, 1041); 4954 Paleoceanography: Sea surface temperature. Received 18 November 2005; Revised 22 May 2006; Accepted 14 June 2006; Published 27 September 2006. G 3 G 3 Geochemistry Geophysics Geosystems Published by AGU and the Geochemical Society AN ELECTRONIC JOURNAL OF THE EARTH SCIENCES Geochemistry Geophysics Geosystems Technical Brief Volume 7, Number 9 27 September 2006 Q09011, doi:10.1029/2005GC001189 ISSN: 1525-2027 Click Here for Full Articl e Copyright 2006 by the American Geophysical Union 1 of 13
Transcript
An aquaculture-based method for calibrated bivalve isotopepaleothermometry
Alan D. Wanamaker Jr., Karl J. Kreutz, Harold W. Borns Jr., and Douglas S. IntroneClimate Change Institute and Department of Earth Sciences, University of Maine, 5790 Bryand Global Sciences Center,Orono, Maine 04469, USA ([email protected])
Scott FeindelDarling Marine Center, University of Maine, Walpole, Maine 04573, USA
Bruce J. BarberSchool of Marine Sciences, University of Maine, Orono, Maine 04469, USA
Now at Galbraith Marine Science Laboratory, Eckerd College, St. Petersburg, Florida 33711, USA
[1] To quantify species-specific relationships between bivalve carbonate isotope geochemistry (d18Oc) andwater conditions (temperature and salinity, related to water isotopic composition [d18Ow]), an aquaculture-based methodology was developed and applied to Mytilus edulis (blue mussel). The four-by-three factorialdesign consisted of four circulating temperature baths (7, 11, 15, and 19!C) and three salinity ranges (23,28, and 32 parts per thousand (ppt); monitored for d18Ow weekly). In mid-July of 2003, 4800 juvenilemussels were collected in Salt Bay, Damariscotta, Maine, and were placed in each configuration. The sizedistribution of harvested mussels, based on 105 specimens, ranged from 10.9 mm to 29.5 mm with a meansize of 19.8 mm. The mussels were grown in controlled conditions for up to 8.5 months, and apaleotemperature relationship based on juvenile M. edulis from Maine was developed from animalsharvested at months 4, 5, and 8.5. This relationship [T!C = 16.19 (±0.14) ! 4.69 (±0.21) {d18Oc VPBD !d18Ow VSMOW} + 0.17 (±0.13) {d18Oc VPBD ! d18Ow VSMOW}2; r2 = 0.99; N = 105; P < 0.0001] isnearly identical to the Kim and O’Neil (1997) abiogenic calcite equation over the entire temperature range(7–19!C), and it closely resembles the commonly used paleotemperature equations of Epstein et al. (1953)and Horibe and Oba (1972). Further, the comparison of the M. edulis paleotemperature equation with theKim and O’Neil (1997) equilibrium-based equation indicates that M. edulis specimens used in this studyprecipitated their shell in isotopic equilibrium with ambient water within the experimental uncertainties ofboth studies. The aquaculture-based methodology described here allows similar species-specific isotopepaleothermometer calibrations to be performed with other bivalve species and thus provides improvedquantitative paleoenvironmental reconstructions.
Wanamaker, A. D., Jr., K. J. Kreutz, H. W. Borns Jr., D. S. Introne, S. Feindel, and B. J. Barber (2006), Anaquaculture-based method for calibrated bivalve isotope paleothermometry, Geochem. Geophys. Geosyst., 7, Q09011,doi:10.1029/2005GC001189.
1. Introduction
[2] Oxygen isotopic analysis of marine biogeniccarbonates (d18Oc) is a standard paleoceanographicmethod used to reconstruct seawater temperatureand/or changes in the isotopic composition of sea-water (d18Ow), when other independent methods canconstrain either water temperature or salinity(Mg/Ca ratios, alkenones, etc.). d18Oc is a functionof seawater temperature [Urey, 1947; Epstein et al.,1953; Craig, 1965; O’Neil et al., 1969], isotopiccomposition of the seawater (related to salinity)[Emiliani, 1966; Shackleton, 1967], and any spe-cies-specific fractionation that occurs during biomi-neralization [Erez, 1978; Shackleton et al., 1973;Swart, 1983; Gonzalez and Lohmann, 1985;McConnaughey, 1989a, 1989b; Owen et al.,2002a; Lorrain et al., 2004]. Substantial informa-tion about marine paleoenvironments can be elicitedfrom stable isotope profiles (d18Oc) from living andfossil bivalves [e.g.,Williams et al., 1982; Arthur etal., 1983; Krantz et al., 1987; Romanek et al., 1987;Wefer and Berger, 1991; Weidman et al., 1994;Klein et al., 1997; Purton and Brasier, 1999; Ivanyet al., 2003; Schone et al., 2004; Carre et al., 2005].However, several factors have been recognized thatcomplicate the understanding of biogenic carbo-nates, which have been described, in part, by previ-ous workers. These factors include carbonateprecipitation in equilibrium with ambient water[Shackleton et al., 1973; McConnaughey, 1989b],pH effects [Spero et al., 1997; Zeebe et al., 2003],ontogeny [Bijma et al., 1998], diagenesis [e.g.,Grossman et al., 1993], seasonal timing and dura-tion of shell growth, and large scale geographictrends in temperature and productivity gradientson shell growth [Jones, 1981; Harrington, 1989;Goodwin et al., 2001;Owen et al., 2002b; Schone etal., 2003, 2005; Goodwin et al., 2004; De Ridder etal., 2004].
[3] In the past, the interpretation of d18Oc has beenbased upon theoretical studies of chemical equilib-rium and kinetics [Urey, 1947; Usdowski andHoefs, 1993], or laboratory experiments involvinginorganic precipitation of CaCO3 from solution[e.g., McCrea, 1950; O’Neil et al., 1969; Tarutaniet al., 1969; Kim and O’Neil, 1997; Zhou andZheng, 2003]. Other methods have employed anempirical calibration of bivalves, done by measur-
ing d18Oc of collected shells from the natural settingand/or from shells grown in a controlled setting andby measuring or estimating d18Ow [Epstein et al.,1953; Craig, 1965; Horibe and Oba, 1972;Grossman and Ku, 1986; Owen et al., 2002a;Chauvaud et al., 2005]. However, previous oxy-gen isotope bivalve calibrations have one or severalpotential limitations: (1) Estimates of temperatureand/or d18Ow were used in the development ofpaleotemperature equations; (2) a limited numberof environmental conditions, such as a single tem-perature or a single salinity, were utilized duringculturing; (3) a limited number of bivalves (as few asone) were grown at a particular temperature andsalinity range; and (4) a limited suite of bivalvespecies was used.
[4] The general isotope calibrations for calcite andaragonite [Epstein et al., 1953; Grossman and Ku,1986] have been applied to a wide variety oforganisms precipitating carbonate skeletons, whichwere not cultured in their calibrations, over time-scales ranging from seasonal to glacial/interglacial.Also, these isotope calibrations were not designed toassess factors such as ‘‘life processes’’ that compli-cate the interpretation of d18Oc. Recognition of theselimitations has led to the development of aquacul-ture-based techniques for selected foraminifera spe-cies [e.g., Erez and Luz, 1983; Spero and Lea, 1993,1996; Bemis et al., 1998; Bijma et al., 1998],bivalves [e.g., Owen et al., 2002a], and corals[e.g., Al-Horani et al., 2003]. We present here amethodology for growing a wide range of bivalvespecies in which key environmental conditions arewell constrained and monitored, and represent rea-sonable growing conditions similar to the naturalenvironment. Our goal is to develop a reproducibleaquaculture-based method for use with a wide rangeof bivalve species, particularly from mid to highlatitudes, where few high-resolution (seasonal)paleo-oceanographic records exist, that facilitatesthe study of shell chemistry, including biologi-cally induced effects as a function of growingconditions.
2. Materials and Methods
2.1. Experimental Bivalve
[5] The relative abundance in coastal environmentsand broad geographical distribution of the intertidal
GeochemistryGeophysicsGeosystems G3G3 wanamaker et al.: bivalve isotope paleothermometry 10.1029/2005GC001189
2 of 13
bivalve M. edulis makes it an ideal species forpaleo-environmental reconstructions. M. edulis hasa current geographic range that extends fromGreenland to North Carolina in the western Atlan-tic Ocean [Wells and Gray, 1960; Read andCumming, 1967]. M. edulis occurs on the eastand west coasts of South America, the FalklandIslands, and along the European coasts from thewestern border of the Kara Sea south to theMediterranean [Tebble, 1966; Seed and Suchanek,1992], and fossils are found in many late-glacialsediments in the circum-Arctic. It is absent fromthe Pacific coast of North America [Seed andSuchanek, 1992]. The southern distribution of thisspecies appears to be limited by an inability totolerate water temperatures exceeding 27!C [Readand Cumming, 1967]. The environmental optimumfor this species is a temperature range of 10–20!C[Bayne et al., 1973] and a variable salinity range of<20 ppt–35 ppt. Because M. edulis is a nearshore-intertidal organism it has the potential to record seasurface temperature (SST) in its shell for a specificcoastal location and time. In addition, it appears tobe an appropriate organism to monitor hydrograph-ic changes over time, because it is found inestuaries and at river mouths. M. edulis is arelatively short-lived organism (6–7 years oldcommon), that deposits annual growth rings [Lutz,1976] and micro-growth rings with tidal and dailyperiodicities [Richardson, 1989]. An adult bluemussel (>2 years) can grow to about 8–10 cm(shell length) allowing for a high-resolution envi-ronmental reconstruction (sub-monthly), and havebeen reported to live up to 18–24 years [Theisen,1973]. Growth rates in their natural setting arevariable, depending on environmental conditions[Incze et al., 1980]. The shell of temperateM. edulisis two layered, with an outer calcitic layer and anaragonitic inside layer [Taylor et al., 1969]. Thearagonitic layer lags the calcitic layer substantially,thus all new growth is calcitic. As the organismcontinues to grow, the aragonitic layer followsoutward toward the mantle.
[6] In mid-July of 2003, 4800 juvenile M. edulis,"15 mm shell length on average, were collectedin Salt Bay, Damariscotta, Maine, USA. Theseanimals were transported to the Darling MarineCenter in Walpole, Maine and were kept moist instorage containers. Animals were sorted to ensurethat a similar distribution of size fractions wereequally distributed in each temperature/salinityconfiguration. Animals were acclimated to theculture temperature gradually for a period of oneweek. On the basis of 100 random samples, the
range was 9.8–20.2 mm, with a mean of 15.3 mm(1s = 2.4 mm).
2.2. Aquaculture Design andImplementation
[7] An aquaculture system was designed at theDarling Marine Center to achieve four temperaturesettings (7, 11, 15 and 19 ± 0.5!C) and threesalinity settings (23, 28, and 32 ± 0.1 ppt). Thisfour by three factorial design allowed 12 differentgrowing conditions to be maintained simultaneously.Each experiment has been duplicated (buckets Aand B). The system consists of three large con-tainers (500-liter) connected to a heating/coolingsystem (Aquanetics Systems), in which four 20-literbuckets were placed into the fresh water bath(Figure 1). The temperature of each bath wasmeasured with a HOBO1 H8 data logger every30 minutes with an accuracy of ±0.5!C (Figure 2),and each HOBO1 H8 data logger was calibrated inan ice-water bath to ensure accuracy. The averagewater exchange in each recirculating bath wasapproximately 10 liters per minute.
[8] Seawater was collected via the flowing seawa-ter laboratory at the Darling Marine Center, andwas pumped from the Damariscotta River at!10 mbelow mean low tide. Seawater was mixed fordesired salinity (23, 28, and 32 ± 0.1 ppt) andstored in 2,460-liter containers and sealed. Salinitymeasurements were made via a YSI model 85oxygen, conductivity, salinity, and temperaturesystem with an accuracy of ±0.1 ppt. We used asimple mixing line based on the mean d18Ow values(see below) of well water and seawater (desiredd18Ow = [0.0029*x] ! 8.6; where x = # of saltwater liters added to the 2,460-liter container) toachieve the desired isotopic composition and sa-linity. Adjustments were made by adding smallvolumes of either well water or seawater to thecontainers to achieve the desired salinity of 23 and28 ± 0.1 ppt. The highest salinity (32 ppt) waslimited by the seasonal cycle in the DamariscottaRiver system, and had a mean oxygen isotopiccomposition (d18Ow) of !1.40% Vienna StandardMean Ocean Water (VSMOW) (N = 8; 1s =0.11%) during June, 2003. For 23 and 28 pptmixtures, seawater was mixed with well water fromthe Darling Marine Center, which had a meand18Ow of !8.62% (VSMOW) (N = 12; 1s =0.14%) during June, 2003. The above procedurewas repeated in October, 2003 to replenish seawa-ter that was used through March, 2004 (meanseawater d18Ow of !1.46% VSMOW; N = 10;
GeochemistryGeophysicsGeosystems G3G3 wanamaker et al.: bivalve isotope paleothermometry 10.1029/2005GC001189wanamaker et al.: bivalve isotope paleothermometry 10.1029/2005GC001189
3 of 13
1s = 0.08% and mean well water d18Ow of!8.59% VSMOW; N = 10; 1s = 0.09%).
[9] Two hundred blue mussels were placed in each20-liter temperature/salinity environment, for atotal of 4800 animals. The animals were culturedfor a total of 8.5 months (mid-July 2003 throughMarch 2004) with five animals being removed
from each configuration monthly for analysis.Ten animals in each temperature and salinity con-figuration were tagged on July 14, 2003 with anumbered shell fish tag ("3 mm) directly adheredto each animal. The shell length for each thesemussels were determined with digital calibers(±0.01 mm) by measuring along the maximum
Figure 2. Temperatures for each of the four freshwater baths are shown with the 8-month mean and standarddeviation. The HOBO1 H8 data logger digitally measured water temperature with an error of ±0.5!C, and theappearance of two lines for each temperature is an artifact of the digital measurement.
Figure 1. Schematic diagram of the experimental design. Each temperature condition is shown vertically, and eachsalinity condition is shown horizontally. Black indicates inflow, and gray indicates outflow. Buckets A and B arereplicates. Buckets to the right of A and B are for water changes. All buckets are in a fresh water bath to maintaindesired temperature setting to within ±0.5!C.
GeochemistryGeophysicsGeosystems G3G3 wanamaker et al.: bivalve isotope paleothermometry 10.1029/2005GC001189
4 of 13
growth axis, and monitored monthly. Averagelinear growth rates (mm/month) were 0.10, 0.08,0.09, and 0.14 for 7!, 11!, 15!, and 19!C temper-ature ranges, respectively. In their natural setting,M. edulis has growth rates of "3 mm/month, withconsiderable variation among individuals [Incze etal., 1980]. Overall, there were no noticeable trendsin growth rate versus salinity or temperature.
[10] Complete water changes for each temperature/salinity environment were made weekly, to removemetabolic waste. The aquaculture design allowedfor one extra 20-liter bucket to be in place withidentical water (isotopic composition and temper-ature) (Figure 1). Mussels were fed twice daily(total of 10 ml) a concentrated spat formula (Inno-vative Aquaculture Products, Ltd.) where 5 ml ofspat was diluted in 1 liter of identical isotopiccomposition water in which they grew. Watersamples for each of the twenty-four buckets werecollected weekly, after water changes were made,to monitor d18Ow (Figure 3). There was no isotopicdifference noted when water was collected prior toand after water changes. Throughout the experi-ment mortality was low (<10%) for all temperatureand salinity configurations for the first five months.Mortality rates were higher ("20–25%) for ani-
mals grown at 20!C for the remainder of theexperiment, while all others remained low.
2.3. Sample Preparation and Analysis
[11] Weekly water samples (d18Ow) were measuredvia a dual-inlet VG/Micromass SIRA, which has along-term precision of ±0.05% (Figure 3). Weeklyd18Ow from each temperature/salinity environmentswere averaged over the growing interval (4, 5, and8.5 months), and used in the isotope calibration(Table A1). All d18Ow (d in % = [(Rsample/Rstandard)! 1] * 1000%; [R = 18O/16O]) values are reportedwith respect to Vienna Standard Mean OceanWater (VSMOW).
[12] Animals were cleaned and air-dried. Shellsamples were further oven dried at 40!C overnight.The periostracum was removed with a razor bladealong the ventral margin. Prior to sampling, eachanimal’s shell length was recorded. The outer edgeof each valve was micro-milled using a variablespeed mounted drill and binocular microscope with6.5x to 40x magnification. To ensure that only newshell carbonate material was used for the isotopecalibration, mussels from each month were sam-pled and standard deviations of isotopic variability(d18Ocalcite ! d18Owater) were calculated. These
Figure 3. The oxygen isotopic composition (d18Ow) for each of the 24 growing environments is shown for a5-month period (only one group of animals were used in the final paleotemperature relationship beyond 5 months,and complete d18Ow values are listed in Table 1). The salinities are 32 ppt (top), 28 ppt (middle), and 23 ppt(bottom). Temperatures are 19!C (red), 15!C (black), 11!C (blue), and 7!C (green). For the replicate environments,the solid lines represent bucket A, while the dashed line represent bucket B. The average standard deviations ford18Ow are (1s) = 0.18%, 0.19%, and 0.17% for 32 ppt, 28 ppt, and 23 ppt, respectively. The observed variabilitiesare 3–4 times larger than the analytical error of the measurement and may be caused by the addition of food/waterthat was isotopically different than the growing conditions, although it was made from the same salinity water. Someevaporation from buckets also could account for a small amount of the variability.
GeochemistryGeophysicsGeosystems G3G3 wanamaker et al.: bivalve isotope paleothermometry 10.1029/2005GC001189
5 of 13
standard deviations (1s) for months 1 and 2 wereon the order of 0.32%–0.52%; thus these animalswere not used in the study. Standard deviations ofisotopic variability (d18Ocalcite ! d18Owater) formonths 4, 5, and 8.5 were on the order of 1s =0.09%–0.12%, and remained constant. In addi-tion, growth rate data (Table A1) was used toestimate how much linear shell material could beremoved during sampling, without mixing previ-ously grown shell with controlled growth. X-raydiffraction was performed on a limited number ofsamples from shell edges to rule out a mixedmatrix of calcite and aragonite. This method isextremely accurate in detecting the presence ofpolymorphs, because of the return signal generatedfrom the different crystal habits of calcite(hexagonal) and aragonite (orthorhombic) duringthe X-ray diffraction analysis. All measurementsindicate that there was no aragonite present nearthe ventral margin, and furthermore by visuallyinspecting the shells the pearly aragonitic layerlagged the calcitic outer shell layer substantially(>1 cm). Shell carbonate analysis (d18Oc) wasperformed on a dual-inlet VG/Micromass Prism,via a 30-place carousel and common acid bathwithout chromium oxide (CrO3) at 90!C, whichhas a long-term precision of ±0.10%. Averageshell samples weighed approximately 100 mg.Samples were calibrated using NBS-19 standardsat the beginning and end of each run, with astandard to sample ratio of 1:3. All shell carbonatevalues (d18Oc) are reported with respect to ViennaPee-Dee Belemnite (VPBD).
2.4. Calibration of Temperature andD18O Relationships
[13] Least squares regression was used to generatethe M. edulis paleotemperature relationship. Rootmean squared errors (RMSE) were calculated at the95% confidence interval (C.I.), and quoted errors onthe slope and intercepts are reported at the 95% C.I.Our shell data (d18Oc) are reported against theinternational VPBD scale and our water data(d18Ow) are reported against the internationalVSMOW scale, which minimizes approximationsand multiple corrections. However, in order tocompare our results to the Epstein et al. [1953],Horibe and Oba [1972] and Kim and O’Neil [1997]calcite equations, corrections had to be made to eachof their data sets or equation, because Epstein et al.[1953] and Horibe and Oba [1972] report the d18Oc
! d18Ow versus [PDB], while Kim and O’Neil[1997] report the d18Oc ! d18Ow versus [SMOW].The water data of Epstein et al. [1953] was con-
verted to the VSMOW scale using the followingrelationship [Friedman and O’Neil, 1977]:
d18Ow VSMOW# $ % 1:00022 * d18Ow PDB# $ & 0:22;
and least squares regression of their data yielded apaleotemperature relationship in the followingform:
Similarly, the calcite equation of Horibe and Oba[1972] was converted to the VPDB - VSMOWscale using the following relationship [Friedmanand O’Neil, 1977]:
d18Ow VSMOW# $ % 1:00022 * d18Ow PDB# $ & 0:22;
and the conversion of their equation yielded apaleotemperature relationship in the followingform:
T'C % 16:10! 4:27 d18OcVPBD!
! d18OwVSMOW"
& 0:16 d18OcVPBD!
! d18OwVSMOW"2: #2$
The calcite data (5 mM solution) of Kim andO’Neil [1997] were corrected (+0.25%) to accountfor differences in acid fractionation factors usedin their work (1.01050) and then converted tothe VPDB scale using the following equation[Friedman and O’Neil, 1977]:
d18Oc VPBD# $ % 0:97006 * d18Ow SMOW# $ ! 29:94;
and least squares regression of their data yielded apaleotemperature relationship in the followingform:
Equation (4) is compared to equations (1), (2),and (3) (Figure 4). M. edulis (equation (4)) isslightly offset relative to the Kim and O’Neil [1997](equation (3)) abiogenic calcite equation over theentire temperature range (7–19!C), and it closelyresembles the commonly used paleotemperatureequations of Epstein et al. [1953] (equation (1)) andHoribe and Oba [1972] (equation (2)). Thecomparison of the M. edulis paleotemperatureequation with the Kim and O’Neil [1997] equili-brium-based model indicates that M. edulis speci-mens used in this study precipitated their shell inisotopic equilibrium with ambient water within theexperimental uncertainties of both studies (Kim andO’Neil [1997] (RMSE ± 0.72!C) and this study(RMSE ± 0.54!C)) (Figure 4).
[15] There is similar isotopic variability (d18Oc !d18Ow) for M. edulis over the upper temperatureranges (1s = 0.12% at 19!C; 0.13% at 15!C;0.12% at 11!C), and slightly less at the lowest
temperature (1s = 0.09% at 7!C) (Figure 4). Theobserved variability is slightly higher or within therange of combined random analytical errors forwater and carbonate analyses (±0.11% [Miller andMiller, 1993]). We determined that approximately0.09% of the variability is shell-derived for alltemperature and salinity conditions; however, thisis less than the analytical error during measurementof d18Oc. We attribute the remainder of the vari-ability to minor changes in the isotopic composi-tion of the water during culture. Owen et al.[2002a] reported isotopic variability of Pectenmaximus (Great Scallop) at any one temperatureof 1s = 0.05%–0.18%. The Kim and O’Neil[1997] inorganic calcite experiment yielded isoto-pic variability of 1s = 0.06%, 0.19%, and 0.10%for 40!, 25!, and 10!C, respectively. Epstein et al.[1953] included temperature-controlled conditionsand multiple animals for only 21.5!C and 19!C(Figure 4), where the isotopic variability was 1s[%] = 0.22% and 0.11%, respectively. The isoto-pic variability from Epstein et al. [1953] was equalto, or nearly twice as great as the isotopic variabil-ity noted in this study for M. edulis. All other datafrom Epstein et al. [1953] had only a single bivalvegrown for each temperature range, and estimates oftemperature and d18Ow were used (Figure 4).Unfortunately, it is not possible to assess theisotopic variability (d18Oc - d18Ow) where onlyone bivalve was grown. On the basis of the isotopicvariability noted in M. edulis, it is likely that
Figure 4. The M. edulis (Maine juveniles) paleotemperature relationship (this study; blue line and blue data points)is compared to the calcite equations of Epstein et al. [1953] (red line and red data points), Horibe and Oba [1972](green line), and Kim and O’Neil [1997] (black line and black data points). Standard deviations for this study arereported for each temperature range in which animals grew. The RMSE ± 0.54!C is reported for this study at 95% C.I.(dashed blue lines).
GeochemistryGeophysicsGeosystems G3G3 wanamaker et al.: bivalve isotope paleothermometry 10.1029/2005GC001189
7 of 13
Table
A1.
Sam
ple
Identification,ShellOxygen
andCarbonIsotopic
Values,Salinity,
ShellLength,Water
Isotopic
Composition,MeasuredTem
perature,Predicted
Tem
perature,Tem
perature
Deviation,MonthsGrown,Estim
ated
Growth
Rates,andEstim
ated
New
LinearGrowth
a
Sam
ple
I.D.
d18O
(VPDB)
d13C
(VPDB)
Salinity,
ppt
Shell
Length,mm
d18O
(VSMOW)
Calcite
-Water
(VPDB
-VSMOW)
Measured
Tem
perature,
!C
Predicted
Tem
perature,
!C
Tem
perature
Deviation,
!CMonths
Grown
Estim
ated
Growth
Rate,
mm/m
onth
Estim
ated
New
Linear
Growth,mm
8-33A-3/29
0.7512
!0.4244
32
12.47
!1.30
2.05
7.10
7.32
!0.22
8.5
0.10
0.85
8-33A-3/29
0.6513
!0.9404
32
10.87
!1.30
1.95
7.10
7.71
!0.61
8.5
0.10
0.85
8-33A-3/29
0.6929
!0.2974
32
19.06
!1.30
1.99
7.10
7.55
!0.45
8.5
0.10
0.85
8-33A-3/29
0.7382
!0.3144
32
15.57
!1.30
2.04
7.10
7.37
!0.27
8.5
0.10
0.85
8-33A-3/29
0.7132
!0.6304
32
14.21
!1.30
2.01
7.10
7.47
!0.37
8.5
0.10
0.85
8-33A-3/29
0.9392
!0.9854
32
21.93
!1.30
2.24
7.10
6.60
0.50
8.5
0.10
0.85
8-33A-3/29
0.9022
!0.7584
32
21.14
!1.30
2.20
7.10
6.74
0.36
8.5
0.10
0.85
8-33A-3/29
0.8692
!0.8074
32
21.10
!1.30
2.17
7.10
6.87
0.23
8.5
0.10
0.85
8-33A-3/29
0.7732
!1.0334
32
20.73
!1.30
2.07
7.10
7.24
!0.14
8.5
0.10
0.85
8-33A-3/29
0.7262
!0.7174
32
20.86
!1.30
2.03
7.10
7.42
!0.32
8.5
0.10
0.85
8-33A-3/29
0.5692
!0.7894
32
21.99
!1.30
1.87
7.10
8.03
!0.93
8.5
0.10
0.85
8-33A-3/29
0.6572
!0.5314
32
15.99
!1.30
1.96
7.10
7.69
!0.59
8.5
0.10
0.85
8-33A-3/29
0.7232
!0.3384
32
20.81
!1.30
2.02
7.10
7.43
!0.33
8.5
0.10
0.85
8-33A-3/29
0.7782
!0.6664
32
17.38
!1.30
2.08
7.10
7.22
!0.12
8.5
0.10
0.85
8-33A-3/29
0.7712
!0.8464
32
19.30
!1.30
2.07
7.10
7.24
!0.14
8.5
0.10
0.85
11-12-8-33B
0.3850
!0.8083
32
20.35
!1.64
2.03
7.39
7.42
!0.03
4.0
0.10
0.40
8-33-A
0.5450
!0.3730
32
21.41
!1.62
2.17
7.39
6.88
0.51
5.0
0.10
0.50
8-33-A
0.4340
!0.6540
32
16.53
!1.62
2.05
7.39
7.31
0.08
5.0
0.10
0.50
8-28-A
!0.8160
!2.0780
28
18.47
!2.82
2.00
7.39
7.50
!0.11
5.0
0.10
0.50
8-28-A
!0.7630
!2.1500
28
17.09
!2.82
2.06
7.39
7.30
0.09
5.0
0.10
0.50
8-28-B
!0.8550
!1.6770
28
19.86
!2.80
1.95
7.39
7.74
!0.35
5.0
0.10
0.50
8-28-B
!0.6050
!1.1550
28
18.66
!2.80
2.20
7.39
6.77
0.62
5.0
0.10
0.50
8-28-B
!0.6230
!1.4860
28
17.04
!2.80
2.18
7.39
6.84
0.55
5.0
0.10
0.50
8-23-A
!1.6300
!2.3710
23
19.88
!3.70
2.07
7.39
7.25
0.14
5.0
0.10
0.50
11-12-12-33A
!0.7110
!1.4423
32
16.77
!1.64
0.93
10.93
11.92
!0.99
4.0
0.08
0.32
11-12-12-33A
!0.5510
!1.4453
32
17.29
!1.64
1.09
10.93
11.23
!0.30
4.0
0.08
0.32
11-12-12-33A
!0.5490
!1.5183
32
15.77
!1.64
1.09
10.93
11.22
!0.29
4.0
0.08
0.32
11-12-12-33B
!0.5370
!1.7950
32
19.83
!1.56
1.02
10.93
11.51
!0.58
4.0
0.08
0.32
11-12-12-33B
!0.4220
!1.7980
32
19.54
!1.56
1.14
10.93
11.02
!0.09
4.0
0.08
0.32
11-12-12-33A
!0.4190
!1.4413
32
16.22
!1.64
1.22
10.93
10.67
0.26
4.0
0.08
0.32
11-12-12-28A
!1.7880
!2.1310
28
19.50
!2.76
0.97
10.93
11.73
!0.80
4.0
0.08
0.32
11-12-12-28A
!1.7030
!2.1660
28
19.15
!2.76
1.06
10.93
11.37
!0.44
4.0
0.08
0.32
11-12-12-28B
!1.5890
!2.4893
28
15.96
!2.79
1.20
10.93
10.76
0.17
4.0
0.08
0.32
11-12-12-28B
!1.5740
!2.2553
28
17.65
!2.79
1.22
10.93
10.69
0.24
4.0
0.08
0.32
11-12-12-23B
!2.6750
!2.4513
23
18.44
!3.92
1.25
10.93
10.57
0.36
4.0
0.08
0.32
11-12-12-23A
!2.5880
!3.4520
23
17.75
!3.93
1.34
10.93
10.17
0.76
4.0
0.08
0.32
11-12-12-23B
!2.5700
!2.9613
23
21.04
!3.92
1.35
10.93
10.14
0.79
4.0
0.08
0.32
11-12-12-23A
!2.5170
!3.4920
23
19.32
!3.93
1.41
10.93
9.87
1.06
4.0
0.08
0.32
12-28A
!1.2160
!1.9425
28
18.04
!2.32
1.10
11.09
11.17
!0.08
5.0
0.08
0.40
GeochemistryGeophysicsGeosystems G3G3 wanamaker et al.: bivalve isotope paleothermometry 10.1029/2005GC001189
8 of 13
Table
A1.(continued)
Sam
ple
I.D.
d18O
(VPDB)
d13C
(VPDB)
Salinity,
ppt
Shell
Length,mm
d18O
(VSMOW)
Calcite
-Water
(VPDB
-VSMOW)
Measured
Tem
perature,
!C
Predicted
Tem
perature,
!C
Tem
perature
Deviation,
!CMonths
Grown
Estim
ated
Growth
Rate,
mm/m
onth
Estim
ated
New
Linear
Growth,mm
12-28A
!1.0580
!1.5245
28
18.53
!2.32
1.26
11.09
10.50
0.59
5.0
0.08
0.40
12-28A
!1.1670
!1.6345
28
15.01
!2.32
1.15
11.09
10.96
0.13
5.0
0.08
0.40
12-28A
!1.1740
!1.8715
28
12.76
!2.32
1.15
11.09
10.99
0.10
5.0
0.08
0.40
12-28A
!1.1980
!1.4085
28
20.02
!2.32
1.12
11.09
11.09
0.00
5.0
0.08
0.40
12-28A
!1.2980
!2.3275
28
21.58
!2.32
1.02
11.09
11.52
!0.43
5.0
0.08
0.40
12-28A
!1.2050
!2.0245
28
22.22
!2.32
1.12
11.09
11.12
!0.03
5.0
0.08
0.40
12-28A
!1.1710
!1.7605
28
18.62
!2.32
1.15
11.09
10.98
0.11
5.0
0.08
0.40
12-28A
!1.1578
!1.5245
28
19.52
!2.32
1.16
11.09
10.92
0.17
5.0
0.08
0.40
12-28A
!1.1560
!1.6345
28
19.51
!2.32
1.16
11.09
10.91
0.18
5.0
0.08
0.40
12-28A
!1.3830
!2.1595
28
20.88
!2.32
0.94
11.09
11.88
!0.79
5.0
0.08
0.40
12-28A
!1.0000
!1.5355
28
18.34
!2.32
1.32
11.09
10.26
0.83
5.0
0.08
0.40
12-28A
!1.2650
!1.7855
28
18.61
!2.32
1.06
11.09
11.38
!0.29
5.0
0.08
0.40
12-28A
!1.1890
!1.8435
28
18.09
!2.32
1.13
11.09
11.05
0.04
5.0
0.08
0.40
11-12-16-33B
!1.4970
!2.4523
32
20.68
!1.49
!0.01
15.18
16.12
!0.94
4.0
0.09
0.36
11-12-16-33A
!1.3070
!1.8060
32
25.46
!1.61
0.30
15.18
14.69
0.49
4.0
0.09
0.36
11-12-16-33B
!1.2550
!1.8273
32
15.47
!1.49
0.24
15.18
15.00
0.18
4.0
0.09
0.36
11-12-16-33A
!1.1900
!1.8910
32
28.49
!1.61
0.42
15.18
14.16
1.02
4.0
0.09
0.36
16-33A
!1.2610
!1.9840
32
21.90
!1.35
0.09
15.18
15.68
!0.50
5.0
0.09
0.45
16-33A
!1.0550
!1.4100
32
14.67
!1.35
0.30
15.18
14.73
0.45
5.0
0.09
0.45
16-33A
!0.9840
!1.7510
32
29.53
!1.35
0.37
15.18
14.41
0.77
5.0
0.09
0.45
16-33A
!1.1750
!1.8280
32
15.71
!1.35
0.18
15.18
15.28
!0.10
5.0
0.09
0.45
16-33A
!1.1710
!1.8470
32
26.25
!1.35
0.18
15.18
15.26
!0.08
5.0
0.09
0.45
16-33A
!1.0300
!1.7540
32
27.19
!1.35
0.32
15.18
14.62
0.56
5.0
0.09
0.45
16-33A
!1.1460
!1.7870
32
18.64
!1.35
0.20
15.18
15.15
0.03
5.0
0.09
0.45
16-33A
!1.2380
!1.8190
32
17.91
!1.35
0.11
15.18
15.57
!0.39
5.0
0.09
0.45
16-33A
!1.2200
!1.8700
32
12.99
!1.35
0.13
15.18
15.49
!0.31
5.0
0.09
0.45
16-33A
!1.1280
!1.8640
32
24.04
!1.35
0.22
15.18
15.06
0.12
5.0
0.09
0.45
16-33A
!1.1880
!1.7500
32
25.79
!1.35
0.16
15.18
15.34
!0.16
5.0
0.09
0.45
16-33A
!1.2440
!2.1370
32
28.59
!1.35
0.11
15.18
15.60
!0.42
5.0
0.09
0.45
16-33A
!1.1380
!2.1340
32
19.06
!1.35
0.21
15.18
15.11
0.07
5.0
0.09
0.45
16-33A
!1.3100
!1.9090
32
22.38
!1.35
0.04
15.18
15.91
!0.73
5.0
0.09
0.45
16-33A
!1.1970
!1.7190
32
23.32
!1.35
0.15
15.18
15.38
!0.20
5.0
0.09
0.45
11-12-16-28B
!2.6870
!2.8530
28
20.74
!2.75
0.06
15.18
15.80
!0.62
4.0
0.09
0.36
11-12-16-28B
!2.6890
!3.5800
28
21.12
!2.75
0.06
15.18
15.81
!0.63
4.0
0.09
0.36
11-12-16-28A
!2.6690
!2.9303
28
27.73
!2.73
0.06
15.18
15.81
!0.63
4.0
0.09
0.36
11-12-16-28A
!2.4880
!3.3393
28
19.88
!2.73
0.24
15.18
14.97
0.21
4.0
0.09
0.36
11-12-16-23B
!3.5400
!3.4930
23
23.31
!3.93
0.39
15.18
14.30
0.88
4.0
0.09
0.36
11-12-16-23B
!3.5337
!3.6550
23
20.44
!3.93
0.40
15.18
14.27
0.91
4.0
0.09
0.36
11-12-16-23A
!3.5037
!2.7810
23
22.99
!3.98
0.48
15.18
13.91
1.27
4.0
0.09
0.36
20-33A
!2.0720
!2.0990
32
23.85
!1.31
!0.76
19.30
19.77
!0.47
5.0
0.14
0.70
20-33A
!1.9020
!2.1300
32
21.66
!1.31
!0.59
19.30
18.93
0.37
5.0
0.14
0.70
GeochemistryGeophysicsGeosystems G3G3 wanamaker et al.: bivalve isotope paleothermometry 10.1029/2005GC001189
GeochemistryGeophysicsGeosystems G3G3 wanamaker et al.: bivalve isotope paleothermometry 10.1029/2005GC001189
10 of 13
similar or greater variability would have beennoted if Epstein et al. [1953] had multiple animalsat all temperature ranges.
[16] The size (shell length) distribution of mussels,based on 105 animals harvested in the experiment,ranged from 10.9 mm to 29.5 mm with a mean sizeof 19.8 mm. This variation in shell length (relatedto the age of the animal) allowed for quantificationof potential vital effects. A comparison of shelllength and temperature deviation (measured tem-perature minus predicted temperature [predictedtemperatures from this study]) is made to deter-mine if there is any shell length-related isotopedisequilibrium. There is a weak positive correlation(r2 = 0.03) between the shell length of the animaland temperature deviation, but the relationship isnot statistically significant. This result suggests thatM. edulis did not exhibit age/size-related disequi-librium during biomineralization over the cultureperiod.
4. Improvements and Future Work
[17] Improvements in this aquaculture-based sys-tem that are being considered focus on improvingconstraints on growth rates, including bio-marking[Kaehler and McQuaid, 1999; Day et al., 1995;Pirker and Schiel, 1993], entire batch measuring,and tagging or etching every individual animal. Inaddition, because mortality rates were relativelylow, it is likely that fewer animals can be grown.Because most of the animals used in this studywere juveniles (less than 2 years-old), we arecurrently culturing adults to further refine theM. edulis paleotemperature relationship. Ongoingwork includes growing M. edulis juveniles andadults from western Greenland to determine ifthere are any large-scale geographic trends in shellcarbonate as a function of growing conditions.Other work may include similar aquaculture-basedexperimentation to evaluate the effects of temper-ature and salinity on trace metal uptake in bivalveshell carbonate, provided the trace element ratios inthe water can be adequately controlled. A potentialbenefit of this work would be to eliminate anunknown in the paleotemperature relationship, thusallowing a well-constrained paleoenvironmentalreconstruction to be made.
5. Summary
[18] During this study we have addressed severallimitations associated with past aquaculture-basedisotope calibrations, including the following:
(1) Precisely measured water temperature andd18Ow values were used in the development of apaleotemperature equation; (2) a wide range ofsalinity and temperatures were utilized duringculture; (3) multiple bivalves (23–28) were grownat each temperature to assess shell isotopic vari-ability; and (4) a species-specific bivalve isotopepaleothermometer was developed. The relation-ships among water temperature, shell carbonate(d18Oc), and water isotopic composition (d18Ow)have been thoroughly examined for M. edulis,hence we are confident in using this bivalve forreconstructing paleoenvironments. Still, past watertemperatures are unknown, and values for theoxygen isotopic composition of ocean waters,especially coastal zones, is not well-constrainedand need to be estimated [e.g., Rye and Sommer,1980]. If d18Ow can be estimated, or determinedindependently, this species-specific aquaculture-based methodology can improve environmentalreconstructions. Further, this experimental designoffers the opportunity to assess many growth-related isotope effects (age, growth rates, lifeprocesses, etc.) in a relatively short time, and todetermine if shell carbonate is precipitated inequilibrium with ambient water.
Appendix A
[19] Additional information is provided (Table A1)including all shell data used in this study (isotopicvalues, shell length, growth), culture conditions(isotopic composition of water, salinity, tempera-ture), as well as predicted temperatures based onthe paleotemperature relationship for M. edulis.
Acknowledgments
[20] We thank Paul Rawson (University of Maine, School ofMarine Sciences) for his advice on mussel cultivation, NancyRaymond and Zachary von Hasseln (University of Mainestudents) for help maintaining the aquaculture systems,Timothy Miller (Darling Marine Center) for help with logisticsand space, and Marty Yates for X-ray diffraction analysis(University of Maine, Earth Sciences). We also thankMary Elliot and an anonymous reviewer for their insightfuland constructive comments that improved this manuscript.This research was funded through the National Science Foun-dation (NSF ATM-0222553).
References
Al-Horani, F. A., S. M. Al-Moghrabi, and D. de Beer (2003),Microsensor study of photosynthesis and calcification in thescleractinian coral, Galaxea fascicularis: Active internal car-bon cycle, J. Exp. Mar. Biol. Ecol., 288, 1–15.
GeochemistryGeophysicsGeosystems G3G3 wanamaker et al.: bivalve isotope paleothermometry 10.1029/2005GC001189
11 of 13
Arthur, M. A., D. F. Williams, and D. S. Jones (1983), Seaso-nal temperature-salinity changes and thermocline develop-ment in the mid-Atlantic Bight as recorded by the isotopiccomposition of bivalves, Geology, 11, 655–659.
Bayne, B. L., R. J. Thompson, and J. Widdows (1973), Someeffects of temperature and food on the rate of oxygen con-sumption by Mytilus edulus L, in Effects of Temperature onEctothermic Organisms, edited by W. Wieser, pp. 181–193,Springer, New York.
Bemis, B. E., H. J. Spero, and D. W. Lea (1998), Reevaluationof the oxygen isotopic composition of planktonic foramini-fera: Experimental results and revised paleotemperatureequations, Paleoceanography, 13, 150–160.
Bijma, J., C. Hemleben, B. T. Huber, H. Erlenkeuser, andD. Kroon (1998), Experimental determination of the ontoge-netic stable isotope variability in two morphotypes of Glo-bigerinella siphonifera (d’Orbigny), Mar. Micropaleontol.,35(2), 141–160.
Carre, M., I. Bentaleb, D. Blamart, N. Ogle, F. Cardenas,S. Zevallos, R. M. Kalin, L. Ortlieb, and M. Fontugne(2005), Stable isotopes and sclerochronology of the bivalveMesodesma donacium: Potential application to Peruvian pa-leoceanographic reconstructions, Palaeogeogr. Palaeclima-tol. Palaeoecol., 228, 4–25.
Chauvaud, L., A. Lorrain, R. B. Dunbar, Y. Paulet,G. Thouzeau, F. Jean, J. Guarini, and D. Mucciarone(2005), Shell of the Great Scallop Pecten maximus as ahigh-frequency archive of paleoenvironmental changes,Geochem. Geophys. Geosyst., 6, Q08001, doi:10.1029/2004GC000890.
Craig, H. (1965), Measurement of oxygen isotope paleotem-peratures, in Stable Isotopes in Oceanographic Studies andPaleotemperatures, edited by E. Tongiori, pp. 161–182,Cons. Naz. delle Ric., Spoleto, Italy.
Day, R. W., M. C. Williams, and G. P. Hawkes (1995), Acomparison of fluorochromes for marking abalone shells,Mar. Fresh. Res., 46, 599–605.
De Ridder, F., R. Pintelon, J. Schoukens, D. P. Gillikin,L. Andre, W. Baeyens, A. de Brauwere, and F. Dehairs(2004), Decoding nonlinear growth rates in biogenic envir-onmental archives, Geochem. Geophys. Geosyst., 5, Q12015,doi:10.1029/2004GC000771.
Emiliani, C. (1966), Isotopic paleotemperatures, Science, 154,851–857.
Epstein, S., R. Buchsbaum, H. A. Lowenstam, and H. C. Urey(1953), Revised carbonate-water isotopic temperature scale,Bull. Geol. Soc. Am., 64, 1315–1326.
Erez, J. (1978), Vital effect on stable-isotope composition seenin foraminifera and coral skeletons, Science, 273, 199–202.
Erez, J., and B. Luz (1983), Experimental paleotemperatureequation for planktonic foraminifera, Geochim. Cosmochim.Acta, 47, 1025–1031.
Friedman, I., and J. R. O’Neil (1977). Compilation of stableisotope fractionation factors of geochemical interest, in Dataof Geochemistry, 6th ed., edited by M. Fleisher, U.S. Geol.Surv. Prof. Pap., 776, 1–37.
Gonzalez, L. A., and K. C. Lohmann (1985), Carbon andoxygen isotopic composition of Holocene reef carbonates,Geology, 13, 811–814.
Goodwin, D. H., K.W. Flessa, B. R. Schone, and D. L. Dettman(2001), Cross-calibration of daily growth increments, stableisotope variation, and temperature in the Gulf of Californiabivalve mollusk Chione cortezi: Implications for paleoenvir-onmental analysis, Palaios, 16, 387–398.
Goodwin, D. H., K. W. Flessa, M. A. Tellez-Duarte,D. L. Dettman, B. R. Schone, and G. A. Avila-Serrano
(2004), Detecting time-averaging and spatial mixing usingoxygen isotope variation: A case study, Palaeogeogr. Pa-laeoclimatol. Palaeoecol., 205, 1–21.
Grossman, E. L., and T. L. Ku (1986), Oxygen and carbonisotope fractionation in biogenic aragonite: Temperature ef-fects, Chem. Geol., 59, 59–74.
Grossman, E. L., H.-S. Mii, and T. E. Yancey (1993), Stableisotopes in Late Pennsylvanian brachiopods from the UnitedStates: Implications for Carboniferous paleoceanography,Geol. Soc. Am. Bull., 105, 1284–1296.
Harrington, R. J. (1989), Aspects of growth deceleration inbivalves: Clues to understanding the seasonal d18O andd13C record—A comment on Krantz et al. (1987), Palaeo-geogr. Palaeoclimatol. Palaeoecol., 70, 399–403.
Horibe, Y., and T. Oba (1972), Temperature scales of arago-nite-water and calcite-water systems, Fossils, 23/24, 69–79.
Incze, L. S., R. A. Lutz, and L. Watling (1980), Relationshipsbetween effects of environmental temperature and seston ongrowth and mortality of Mytilus edulis in a temperaturenorthern estuary, Mar. Biol., 57, 147–156.
Ivany, L. C., B. H. Wilkinson, and D. S. Jones (2003), Usingstable isotopic data to resolve rate and duration of growththroughout ontogeny: An example from the surf clam, Spi-sula solidissima, Palaios, 18, 126–137.
Jones, D. S. (1981), Repeating layers in the molluscan shell arenot always periodic, J. Paleontol., 55, 1076–1082.
Kaehler, S., and C. D. McQuaid (1999), Use of the fluoro-chrome calcein as in situ growth marker in the brown musselPerna perna, Mar. Biol., 133, 455–460.
Kim, S. T., and J. R. O’Neil (1997), Equilibrium and none-quilibrium oxygen isotope effects in synthetic carbonates,Geochim. Cosmochim. Acta, 61, 3461–3475.
Klein, R. T., K. C. Lohmann, and G. L. Kennedy (1997),Elemental and isotopic proxies of paleotemperature and pa-leosalinity: Climate reconstruction of the marginal northeastPacific ca. 80 ka, Geology, 25(4), 363–366.
Krantz, D. E., D. F. Williams, and D. S. Jones (1987), Ecolo-gical and paleoenvironmental information using stable iso-tope profiles from living and fossil molluscs, Palaeogeogr.Palaeoclimatol. Palaeoecol., 58, 249–266.
Lorrain, A., Y.-M. Paulet, L. Chauvaud, R. Dunbar,D. Mucciarone, and M. Fontugne (2004), d13C variation inscallop shells: Increasing metabolic carbon contribution withbody size?,Geochim. Cosmochim. Acta, 68(17), 3509–3519.
Lutz, R. A. (1976), Annual growth patterns in the inner shelllayer of Mytilus edulis L, J. Mar. Biol. Assoc. U. K., 56,723–731.
McConnaughey, T. (1989a), 13C and 18O isotopic disequili-brium in biological carbonates: I. Patterns, Geochim. Cos-mochim. Acta, 53, 151–162.
McConnaughey, T. (1989b), 13C and 18O isotopic disequili-brium in biological carbonates: II. In vitro simulation ofkinetic isotope effects, Geochim. Cosmochim. Acta, 53,163–171.
McCrea, J. M. (1950), On the isotopic chemistry of carbonatesand a paleotemperature scale, J. Chem. Phys., 18, 849–857.
Miller, J. C., and J. N. Miller (1993), Statistics for AnalyticalChemistry, Prentice-Hall, Upper Saddle River, N. J.
O’Neil, J. R., R. N. Clayton, and T. K. Mayeda (1969), Oxy-gen isotope fractionation in divalent metal carbonates,J. Chem. Phys., 51(12), 5547–5557.
Owen, R., H. Kennedy, and C. Richardson (2002a), Experi-mental investigation into partitioning of stable isotopes be-tween scallop (Pecten maximus) shell calcite and sea water,Palaeogeogr. Palaeoclimatol. Palaeoecol., 185(1), 163–174.
GeochemistryGeophysicsGeosystems G3G3 wanamaker et al.: bivalve isotope paleothermometry 10.1029/2005GC001189
12 of 13
Owen, R., H. Kennedy, and C. Richardson (2002b), Isotopicpartitioning between scallop shell calcite and seawater: Ef-fect of shell growth rate, Geochim. Cosmochim. Acta,66(10), 1727–1737.
Pirker, J. G., and D. R. Schiel (1993), Tetracycline as a fluor-escent shell-marker in the abalone Haliotis iris, Mar. Biol.,116, 81–86.
Purton, L., and M. Brasier (1999), Giant protist Nummulitesand its Eocene environment: Life span and habitat insightsfrom d18O and d13C data from Nummulites and Venericardia,Hampshire basin, UK, Geology, 27(8), 711–714.
Read, K. R., and K. B. Cumming (1967), Thermal toleranceof bivalve molluscs modiolus (L.), Mytilus edulis L., andBrachidontes demissus, Biochem. Physiol., 22, 149–155.
Richardson, C. A. (1989), An analysis of the microgrowthbands in the shell of the common mussel Mytilus edulis,J. Mar. Biol. Assoc. U. K., 69(2), 477–491.
Romanek, C. S., D. S. Jones, D. F. Williams, D. E. Krantz, andR. Radtke (1987), Stable isotopic investigation of physiolo-gical and environmental changes recorded in shell carbonatefrom the giant clam Tridacna maxima, Mar. Biol., 94, 385–393.
Rye, D. M., and M. A. Sommer II (1980), Reconstructingpaleotemperature and paleosalinity regimes with oxygenisotopes, in Skeletal Growth of Aquatic Organisms: Bio-logical Record of Environmental Change, edited by D. C.Rhodes and R. A. Lutz, pp. 169–201, Springer, NewYork.
Schone, B. R., K. Tanabe, D. L. Dettman, and S. Sato (2003),Environmental controls on shell growth rates and d18O of theshallow marine bivalve mollusk Phacosoma japonicum inJapan, Mar. Biol., 142, 473–485.
Schone, B. R., A. D. Freyre Castro, J. Fiebig, S. D. Houk,W. Oscmann, and I. Kroncke (2004), Sea surface water tem-peratures over the period 1884–1983 reconstructed fromoxygen isotope ratios of a bivalve mollusk shell (Arcticaislandica, southern North Sea), Palaeogeogr. Palaeoclima-tol. Palaeoecol., 212, 215–232.
Schone, B. R., S. Houk, A. D. Freyre Castro, J. Fiebig,W. Oschmann, I. Kroncke, W. Dreyer, and F. Gosselck(2005), Daily growth rates in shells of Arctica islandica:Assessing sub-seasonal environmental controls on a long-lived bivalve mollusk, Palaios, 20, 78–92.
Seed, R., and T. H. Suchanek (1992), Population and com-munity ecology of Mytilus, in The Mussel Mytilus: Ecol-ogy, Physiology, Genetics and Culture, vol. 25, edited byE. D. Gossling, pp. 87–170, Elsevier, New York.
Shackleton, N. (1967), Oxygen isotope analyses and Pleisto-cene temperatures re-assessed, Nature, 215, 15–17.
Shackleton, N. J., J. D. H. Wiseman, and H. A. Buckley(1973), Non-equilibrium isotopic fractionation between sea-water and planktonic foraminiferal tests, Nature, 242, 177–179.
Spero, H. J., and D. W. Lea (1993), Intraspecific stable isotopevariability in the planktic foraminifera Globigerinoides sac-
culifer: Results from laboratory experiments, Mar. Micropa-leontol., 22, 221–234.
Spero, H. J., and D. W. Lea (1996), Experimental determina-tion of stable isotope variability in Globigerina bulloides:Implications for paleoceanographic reconstructions, Mar.Micropaleontol., 28, 231–246.
Spero, H. J., J. Bijma, D. W. Lea, and B. E. Bemis (1997),Effect of seawater carbonate concentration on foraminiferalcarbon and oxygen isotopes, Nature, 390, 497–500.
Swart, P. K. (1983), Carbon and oxygen isotope fractionationin scleractinian corals: A review, Earth Sci. Rev., 19, 51–80.
Tarutani, T., R. N. Clayton, and T. K. Mayeda (1969), Theeffect of polymorphism and magnesium substitution on oxy-gen isotope fractionation between calcium carbonate andwater, Geochim. Cosmochim. Acta, 33, 987–996.
Taylor, J. D., W. J. Kennedy, and A. Hall (1969), The shellstructure and mineralogy of the Bivalvia, Bull. Br. Mus., 3,1–125.
Tebble, N. (1966), British Bivalve Seashells, 212 pp., Trusteesof the Br. Natl. Mus. Nat. Hist., London.
Theisen, B. F. (1973), The growth of Mytilus edulis L. (Bival-via) from Disko and Thule district, Greenland, Ophelia, 12,59–77.
Urey, H. C. (1947), The thermodynamic properties of isotopicsubstances, J. Chem. Soc., 1947, 562–581.
Usdowski, E., and J. Hoefs (1993), Oxygen isotope exchangebetween carbonic acid, bicarbonate, carbonate, and water: Are-explanation of the data of McCrea (1950) and an expres-sion for the overall partitioning of oxygen isotopes betweenthe carbonate species and water, Geochim. Cosmochim. Acta,57, 3815–3818.
Wefer, G., and W. H. Berger (1991), Isotope paleontology:Growth and composition of extant calcareous species, Mar.Geol., 100, 207–248.
Weidman, C., G. A. Jones, and K. C. Lohmann (1994), Thelong-lived mollusk Arctica islandica: A new paleoceano-graphic tool for the reconstruction of bottom temperaturesfor the continental shelves of northern Atlantic Ocean,J. Geophys. Res., 99, 18,305–18,314.
Wells, H. W., and I. E. Gray (1960), The seasonal occurrenceofMytilus edulis on the Carolina coast as a result of transportaround Cape Hatteras, Biol. Bull., 119, 550–559.
Williams, D. F., M. A. Arthur, D. S. Jones, and N. Healy-Williams (1982), Seasonality and mean annual sea surfacetemperatures from isotopic and sclerochronological records,Nature, 296, 432–434.
Zeebe, R. E., D. A. Wolf-Gladrow, J. Bijma, and B. Honisch(2003), Vital effects in foraminifera do not compromise theuse of d11B as a paleo-pH indicator: Evidence from model-ing, Paleoceanography, 18(2), 1043, doi:10.1029/2003PA000881.
Zhou, G. T., and Y. F. Zheng (2003), An experimental study ofoxygen isotope fractionation between inorganically precipi-tated aragonite and water at low temperatures, Geochim.Cosmochim. Acta, 67(3), 387–399.
GeochemistryGeophysicsGeosystems G3G3 wanamaker et al.: bivalve isotope paleothermometry 10.1029/2005GC001189
